Vapor Transport Through Fractures and Other High-Permeability Paths: Its Role in the Drift Scale Test at Yucca Mountain, Nevada

Heating unsaturated fractured tuff sets off a series of complicated thermal-hydrological (TH) processes, which result in large-scale redistribution of moisture in the host rock. Moisture redistribution arises from boiling of water near heat sources, transport of vapor away from those heat sources, condensation of that vapor in cooler rock, and subsequent gravity drainage of condensate through fractures. Vapor transport through high-permeability paths, which include both the fractures in the rock and other conduits, contributes to the evolution of these TH processes in two ways. First, the highly permeable natural fractures provide easy passage for vapor away from the heat sources. Second, these fractures and other highly permeable conduits allow vapor (and the associated energy) to escape the rock through open boundaries of the test domain. The overall impact of vapor transport on the evolution of the TH processes can be more easily understood in the context of the Drift Scale Test (DST), the largest ever in situ heater test in unsaturated fractured tuff. The DST, in which a large volume of rock has been heated for four years now, is located in the middle nonlithophysal (Tptpmn) stratigraphic unit of Yucca Mountain, Nevada. The fractured tuff in Tptpmn contains many well-connected fractures. In the DST, heating is provided by nine cannister heaters placed in a five-meter-diameter Heated Drift (HD) and fifty wing heaters installed orthogonal to the axis of the HD. The test has many instrumentation boreholes, some of which are not sealed by packers or grout and may provide passage for vapor and energy. Of these conduits, the boreholes housing the wing heaters are most important for vapor transport because of their proximity to heat sources. While part of the vapor generated by heating moves away from the heat sources through the fractures and condenses elsewhere in the rock, the rest of the vapor, under gas-pressure difference, enters the HD by way of the high-permeability wing heater boreholes and escapes the test block through an open bulkhead that connects the HD to the outside world. We show that this vapor transport makes a significant difference in the validation of numerical models against TH processes in the DST. A huge volume of data, including changes in temperature and saturation of the rock, has been collected from the DST. Sophisticated conceptual and numerical models, based on the TOUGH2 simulator, have been developed to analyze these data and to help develop a better understanding of various aspects of coupled TH processes in unsaturated fractured tuff. In general, these models have predicted a close match between measured and simulated results, indicating a good representation of the underlying physical processes. However, there are subtle differences in the predictions from these models. Of particular interest here are two models: One in which vapor transport was considered through the natural fractures only, and the other in which vapor transport through the boreholes housing the wing heaters was included in addition to that through natural fractures. Direct statistical comparison of simulated and measured temperatures from more than 1,700 sensors yielded a mean error of 3-4^oC for the first model, indicating that less heat was retained in the test block than that predicted by the model. On the other hand, a similar statistical comparison yielded a mean error of 1-2^oC for the second model, suggesting that inclusion of vapor loss through the boreholes produces results closer to the measured data.